0

Suppose I have to predict if a certain product from an assembly line in a factory will be a scrap. This product has let's say 'static' data like a certain shape. A certain vendor, etc. And, it can have 'dynamic' data this meaning it can have for example: one or more sets of measurements (pressures,temperatures ,etc) from production processes.

How to treat this 'dynamic' features ?

Somehow it doesn't seem right to repeat the 'static' data for all 'dynamic' events. And using the mean of 'dynamic' features would dilute the information.

I'm thinking to encode this 'dynamic' data in a similar way phrases with variable number of words are encoded in fix length vectors with lstm networks. What do you think ?

Academic
  • 482
  • 3
  • 13
Toma Dragos
  • 1
  • 1
  • Do you mean that the values for these dynamic features change across time, as opposed to the static features which have a fixed value? Or is it that the number of these features is variable? These dynamic features are available for all the instances, right? – Erwan Nov 12 '20 at 12:16
  • there are not any timestamps. In the training set, each product has at least one set of dynamic features – Toma Dragos Nov 12 '20 at 14:17

0 Answers0